SimCURL: Simple Contrastive User Representation Learning from Command Sequences
This work addresses user modeling for personalized recommendations, but it is incremental as it builds on existing contrastive learning methods with specific adaptations for command sequences.
The paper tackles the problem of learning user representations from unlabeled command sequences, proposing SimCURL, a contrastive self-supervised framework that improves performance on downstream tasks like experience and expertise classification using a dataset of over half a billion commands.
User modeling is crucial to understanding user behavior and essential for improving user experience and personalized recommendations. When users interact with software, vast amounts of command sequences are generated through logging and analytics systems. These command sequences contain clues to the users' goals and intents. However, these data modalities are highly unstructured and unlabeled, making it difficult for standard predictive systems to learn from. We propose SimCURL, a simple yet effective contrastive self-supervised deep learning framework that learns user representation from unlabeled command sequences. Our method introduces a user-session network architecture, as well as session dropout as a novel way of data augmentation. We train and evaluate our method on a real-world command sequence dataset of more than half a billion commands. Our method shows significant improvement over existing methods when the learned representation is transferred to downstream tasks such as experience and expertise classification.